how to locate each holes positions relative to the center of engine shaft?

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

howardyin

Well-Known Member
Joined
May 23, 2020
Messages
54
Hello , guys here ,

how to locate each holes accurate positions relative to the center of engine shaft ?


I am a homebuilder now modifying my Honda L15A , I am wondering how to make sure the shaft center position relative to the holes those used to hold the plate which hold shaft of output gears. (to keep engine shaft Coaxial to belt gear shaft)
r05.jpg
 

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
578
As the plate you are referencing is clearly a product of a cnc machine,why not back track
that part and get whoever made it to use there
files to make the next piece you need?
 

howardyin

Well-Known Member
Joined
May 23, 2020
Messages
54
As the plate you are referencing is clearly a product of a cnc machine,why not back track
that part and get whoever made it to use there
files to make the next piece you need?
thanks. The pic is not mine. I just brought one engine to study , have done nothing yet. thinking about how to measure ( the size ,the location)


what in my hand just in this pic, fisrt thing I need to do is to get the locations of the holes ,then put into CAD software.
微信图片_20210727202935.jpg
 
Last edited:

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
15,449
Location
Memphis, TN
The answer is relative to the tools you have available. Cardboard template and 2 ft machinist ruler will work if you are careful. If you can mill stuff, you can make plugs with a center dimple and put in the crank center and mounting holes and use the ruler or big caliper. You could get fancy with some sort of protractor that could come up with angles when bolted to the crank. Lidar scanner or measuring arms if you have the access to every tool in the world.
I would remove the flywheel and go with the cardboard and ruler. Might take 4-5 tries to be happy, but a lot cheaper than a Lidar setup.
 

PMD

Well-Known Member
Joined
Apr 11, 2015
Messages
508
Location
Martensville SK
Obviously, digital scanning is the ultimate, but there are a few things to keep in mind: If you are measuring from the crankshaft position, it is NOT concentric to the bore. If you are going to scan or even physically measure, you need to knock the engine apart and measure the rear main bearing BORE, not the crank position. Also, remember there are tolerances in the original equipment as to the hole size and location, but there should be two hollow dowels to locate the gearbox, and those should be very close to correct near zero tolerance.

If I were to physically measure, would NOT use centers - as it is extremely difficult to make accurate measurements to the center of a hole or point. I would start by turning an exact fit to main bearing bore piece of shaft and bolting it under the rear main cap sticking out of the back of the engine. Make exact fit pins (or bolts) you can press and/or screw into the pattern of bolt holes. Use a large digital caliper to measure the distance between each and every hole, knowing full well the exact diameter to be able to calculate the virtual centerlines. The most critical of course will be the crank stub to existing trans/bellhousing locating sleeves. Not e:THERE is a big risk, as this will give you two of the critical dimensions, but unless in a straight line through crank center (let's call that the "X" plane) you need to get the "Y" axis from the remaining bolt holes. You can't just shove a random bolt in those holes (see above) or screw one loosely into a threaded hole (use a nut on the thread so it pulls it tight and centered into the threaded hole - and MEASURE the bolt. For instance, a 3/8 bolt is by no means 0.375 in diameter! Nor all that straight or concentric to the thread - so you need to make those measurements SEVERAL times with the bolt locked at different depths of a single thread).

Be careful, take your time, be THOROUGH and aim for zero tolerance - realizing you will have to accept factory tolerance PLUS you own level of error (tolerances are all cumulative). Extremely important if there is going to be a gear drive involved - but if THAT is the case, I hope the gear centers and loads are all going through the gear case, not the crankshaft.

Measure ALL of the distance between shaft/dowel outer diameters, and the radii to centers, then make a drawing using the numbers and they will give you set of center distances that you can then turn into X and Y co-ordinates geometrically. BTW: I have done layouts in this manner to make zero-zero drilled patterns and they will easily bolt together (thanks to bolt diameter actual sizes being less than careful measuring tolerances. Gets a LOT more difficult when dowel pins and gear centers are involved, but still do-able.
 
Last edited:

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,951
Location
Rocky Mountains
If you are building from scratch (?) just use a cardboard template to build the plate leaving the crank hole undersized till last.

The locating dowels PMD mentioned need/should be close tolerance and locate your built part to the engine with no play. Once this is all done bolt the parts to the block - then - bore the crank center hole.
As PMD mentioned the crank has bearing clearance and won't reliably locate the bore. Find a junk engine block with the original trans bolt pattern with at least 2 usable crank bearing saddles and use them to line bore the new drive parts. Mixing ball/roller bearings and hydrostatic bearings on the same shaft introduces another factor to be considered.

Or:
Find a copy of the original dimensions and use them. Sometimes you can get lucky and find the original factory dimensions for very popular engines.
 

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
578
thanks. The pic is not mine. I just brought one engine to study , have done nothing yet. thinking about how to measure ( the size ,the location)


what in my hand just in this pic, fisrt thing I need to do is to get the locations of the holes ,then put into CAD software.
View attachment 113510
Ok then one principal is that each hole you need can be treated as a point,on a circle,centered on your crankshaft.
And then each hole can be measured for distance to each other.
If any three holes line up in such a way to make
right angle triangles,so much the better.
Plenty of other good advice here.
Edit:Just re read the other posts and ,never
considered that the crank center could not be used.So how about finding the crank bearing clearances and wrapping the crank
in shim stock ,thus lockng it and then useing that as a center?
 
Last edited:

pfarber

Well-Known Member
Joined
Feb 21, 2019
Messages
827
Location
Dollywood
Also when prototyping a retainer/plate start with a thin plate (cheaper and faster to mill/cut) and test fit/adjust until you get dimensions that are tight. Don't try nor expect to get it right on the first go. If you do, great, but have the patience to do it a few times and dial it in.

There is a reason every mill has a scrap barrel right next to it.
 

howardyin

Well-Known Member
Joined
May 23, 2020
Messages
54
Also when prototyping a retainer/plate start with a thin plate (cheaper and faster to mill/cut) and test fit/adjust until you get dimensions that are tight. Don't try nor expect to get it right on the first go. If you do, great, but have the patience to do it a few times and dial it in.

There is a reason every mill has a scrap barrel right next to it.
yes, I usualy put it in CAD, then laser cut plywood sample to test fit。
 

Geraldc

Well-Known Member
Joined
Nov 12, 2011
Messages
664
Location
nz
If you are measuring from the crankshaft position, it is NOT concentric to the bore.
But the drive has to be lined up with the crankshaft .
It is the critical part. I would be very surprised if the shaft was not central to the bearing bore. Bearings are either concentric or scrap.
When setting up use a magnetic stand and a dial gauge attached to the end of the crank. Auto bellhousing shown but same for redrive plate.
1627453883205.png
 

cvairwerks

Well-Known Member
Joined
May 12, 2010
Messages
297
Location
North Texas
We used to make quick and dirty hole templates with plexiglass to do repairs. Tape in place and then use a sharp scribe to mark the hole centers. If high accuracy was needed, two holes would be oversized and then we would glue drill bushings in place with guide pins into the structure. About 2 hours later we had a temp drill guide.

Lots of our temp tools ended up being remade into permanent metal tools with all correct dimensions.
 

PMD

Well-Known Member
Joined
Apr 11, 2015
Messages
508
Location
Martensville SK
I have done many trans adapters, prop direct drives and even some PSRUs and can tell you from the school of some very hard knocks that you want to be EXTREMELY accurate and careful in your measurements. AND, extremely aware of the nature of what the crank is doing vs what the drive train behind it will suffer.

Just to give you an example: I spent more than a year trying to find out why a commercial engine (widely assumed to be "bulletproof") would tear the center out of clutches and why a huge, well-regarded gearbox behind it would trash the cluster gear (and the $5k gearbox) with great regularity. It all came back to the TVs in the crank (as a result of a very interesting harmonic condition in the FUEL RAILS!!!!!!) that would pass through (as a timing error) into the driveline. Also, there was some problems with concentricity of the pilot bearing hole in the crank!!!!. What I am saying is assume NOTHING, aim for zero tolerance (that you will never achieve). In this case, the bulletproof engine was firing shots at the transmission and clutch with great accuracy.

In a perfect world (again, one never achieved) I aim to have a flexible coupling between the crank and whatever is next. If a gearbox, ideally the next shaft is fully supported and the connection to the crank can then tolerate the end float of the crank, any radial eccentricities (some shafts even walk around in a cluster of circles within circles). As you can see from an automotive gearbox, that crank end COULD be designed to support one end of a shaft with a gear on it, and they usually are. BUT: if that crank doesn't behave as predicted and/or assumed, it can destroy the gear at the other end in a hurry. Multiply these things 10x when considering the angular velocity and resonances in the crank output and prop/gearbox feedback.

If you plan on making a PSRU that uses the crank to hold a drive pulley or much worse a drive GEAR, the range of tolerances from trying to transfer dimensions to a piece of cardboard, plywood, steel, aluminum....whatever - are far beyond what especially a gear drive will ever tolerate. MEASURE with extreme care and thought. Yes, digital scan extremely good, but as someone already pointed out - if you can get the OEM's actual dimensions (often available...SOMEWHERE) even better.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,236
Location
NJ
if you can get one of these parts then you can transfer the hole locations.......your tolerance for bearings and gears locations should be very close (less than .001"?) which will be very difficult without a CMM (computer measuring machine) or milling machine and dial indicator to dial indicate the locations of the holes. It might be better to buy that plate from someone that makes PSRU for the engine already or bite the bullet and buy the whole PSRU. as all the TV issues should be worked out. As a CAD designer I know there is lots more to this than just cad design.

1627482646137.png
 

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
578
While I concur with the need to have gear sets
aligned in both dimensions to .001 or better,I am left with a quandry as to the value of measuring crank shaft run out when evaluating
a motor.
Or is crank run out only a broad hint, and that hanging anything like a belt drive pulley off of a crank just not a good idea.
The design I am working on has the drive pulley
driven from a splined stub,and everything else
built for easy servicing as well.
I would like to build an belt tensioning adjustment device that has double cut adjuster
screws that have been calibrated to a vernier scale etched or engraved onto the drives frames.
 

blane.c

Well-Known Member
HBA Supporter
Joined
Jun 27, 2015
Messages
5,015
Location
capital district NY
If you have 0.005" clearance at the main journals and a crankshaft 16" long between front and rear main journals and "the flange" you measure from 2" from that, you could easily have 0.006" side to side/top to bottom motion with a "dry" engine but with oil and pressure between the crank and journal bearings it will be much less than that.

The actual path of the crankshaft will not be concentric with the center but toroidal depending on pressures from the cylinders, cam gear, other gears and other miscellaneous like a propeller for instance.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
5,236
Location
NJ
double cut adjuster
words are a poor substitute for drawings and much would depend on the the details of the design....If you are adding a bearing to the crank to support the belt drive that is one design where concentric machining of the bearing mounting at the crank would be necessary.....if not the belt drive probably would not need as an accurate location as the adjuster would take up position in one plane. However the prop mount would have to be square and true to the crank to minimize belt wear or walk. I doubt you would need much more than a fine thread bolt for the adjuster as the tension of the belt is somewhat subjective.
 

PMD

Well-Known Member
Joined
Apr 11, 2015
Messages
508
Location
Martensville SK
if you can get one of these parts then you can transfer the hole locations.......your tolerance for bearings and gears locations should be very close (less than .001"?) which will be very difficult without a CMM (computer measuring machine) or milling machine and dial indicator to dial indicate the locations of the holes.
If you measure to pins from each hole and each other, you end up with a set of numbers that can easily be turned with basic trig into a dead accurate set of x-y co-ordinates. Impossible with something like 3 holes more-or-less in a line, but very easy with a full bolt pattern typical of backside of most engines.
 

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
578
words are a poor substitute for drawings and much would depend on the the details of the design....If you are adding a bearing to the crank to support the belt drive that is one design where concentric machining of the bearing mounting at the crank would be necessary.....if not the belt drive probably would not need as an accurate location as the adjuster would take up position in one plane. However the prop mount would have to be square and true to the crank to minimize belt wear or walk. I doubt you would need much more than a fine thread bolt for the adjuster as the tension of the belt is somewhat subjective.
I will get to drawings,and then start filling the
waste paper basket with those.
Right now I am throwing out ideas, and getting
better ones.
Useing a doughnut/guibo on the flywheel and driving the small pulley will isolate the belt drive from the toroidal motion of the crankshaft and having a spline shafts for
both pulleys and the doughnut will alow for
pulling it apart for inspection and belt changes.
I am designing for a drive shaft to transmit power from the belt drive to the prop.
So the motor and belt drive will see only torque.
 
Top